Power Integrations, Inc. (POWI) Q3 2020 Earnings Report, Transcript and Summary

Good afternoon. My name is [Tegan], and I will be your conference operator today. At this time, I would like to welcome everyone to the Power Integrations Third Quarter Earnings Conference call. [Operator Instructions] Thank you. Mr. Joe Shiffler, you may begin.

Good afternoon, thanks for joining us. With me on the call today are Balu Balakrishnan, President and CEO of Power Integrations and Sandeep Nayyar, our Chief Financial Officer. On August 18, our integrations executed two-for-one stock split in the form of a stock dividend with one new share issued for each outstanding share. In our presentation of the Q3 results, all prior period shares outstanding and per share measures have been adjusted for the split. Also during this call, we will refer to financial measures not calculated according to GAAP. Non-GAAP measures exclude stock-based compensation expenses, amortization of acquisition-related intangible assets and the tax effects of these items. A reconciliation of non-GAAP measures to our GAAP results is included in our press release. Our discussion today, including the Q&A session, will include forward-looking statements denoted by words like will, would, believe, should, expect, outlook, forecast, anticipate, and similar expressions that look toward future events or performance. Such statements are subject to risks and uncertainties that may cause actual results to differ materially from those projected or implied. Such risks and uncertainties are discussed in today's press release and in our Form 10-K filed with the SEC on February 7, 2020. Finally, this call is the property of Power Integrations and any recording or rebroadcast is expressly prohibited without the written consent of Power Integrations. Now I'll turn the call over to Balu.

Thanks, Joe, and good afternoon. Third quarter revenues were $121 million, up 13% sequentially and above the top end of our guidance. We are forecasting healthy sequential growth again in the fourth quarter with revenues of $130 million, plus or minus $5 million. At the midpoint of the range, our full year revenues would be up 11% compared to 2019, well ahead of our expected growth rate for the analog semiconductor industry. Underpinning our strong outlook is a sharp acceleration in orders in August and September as well as a surge in distribution sell through, which outpaced sell-in across all four end market categories in Q3. The differential was widest in the consumer category, driven by strong demand for appliances and in communications where we are benefiting from short and long-term secular trends in the smartphone market. Short-term trend is the dislocation caused by U.S. trade and national security policies, which is enabling our OEM customers to take share in the handset market. The longer-term trend, of course, is the increasing adoption of fast charging in mobile device market and our continued leadership in the fast charging market, thanks to our revolutionary InnoSwitch products and our GaN technology while InnoSwitch ICs are used in a broad range of applications. Their success in high-volume smartphone market has made them the highest selling power supply ICs in the world. And earlier this week, we announced that we have shipped more than 1 billion InnoSwitch chips in the six years since their introduction. Still the only power supply ICs in the market capable of crossing the isolation barrier, InnoSwitch ICs have proven extremely well suited for fast chargers. Due to their high level of integration and energy efficiency, which enable high-power chargers with compact form factors and no heatsinks. We took these capabilities to another level last year with the introduction of GaN-based InnoSwitch products, which are leading the fast-growing market for high-end ultrafast chargers. We expect such chargers to become main-stream as the power requirements of 5G drive power levels higher. And as OEMs take, a more strategic view of charging to capitalize on new technologies like GaN, USB PD and multiport chargers. While some OEMs continue to supply fast chargers with their phones using charging speed as the differentiating feature of their products. Other OEMS, including one that introduced new handsets earlier this month are adopting an accessory model, allowing consumers to choose which charger they want, if any. This is an evolution that we have been - we have seen coming for some time. And while no one can predict exactly how it will play out, it's clear that a fresh wave of innovation in mobile device chargers is coming and we are well prepared for it, thanks to our investments in advanced technologies like InnoSwitch and GaN. Fast charging should be the largest contributor to our growth in Q4 and we anticipate continued success in this area in 2021. While our GaN products have achieved strong uptick in mobile device chargers, we expect technology to be adopted more broadly as customers look to take advantage of its efficiency and as we expand our GaN offerings. In addition to our GaN-based InnoSwitch chips, we have introduced GaN products targeting LED lighting and display applications. And we now have more than 80 customers in production with GaN products across over a dozen applications, including cellphone chargers but also appliances, notebook adapters, TVs and USB wall outlets. We are on track to double our GaN product revenue this year. And based on our pipeline of ongoing design activity, we think there is a good chance it could double again in 2021. Last week, we introduced our latest GaN product called MinE-CAP. MinE-CAP is a normal concept that enables designers to reduce the size of the input capacitors used in adapters, chargers and auxiliary power supplies. These large capacitors account for a significant percentage of the volume of the compact adapter. MinE-CAP enables the use of much smaller capacitors for a given power level resulting in a reduction of up to 40% in the size of the - adapter. As our customers continue to look for new ways to deliver more power with smaller form factors, we think MinE-CAP will prove highly attractive, and it's a great example of the kind of innovation we have unleashed with our GaN technology. And now I'll turn it over to Sandeep.

Thanks, Balu, and good afternoon. As usual, I will focus my remarks primarily on the non-GAAP results which are reconciled to GAAP in our press release tables. Q3 revenues were $121.1 million, up 13% sequentially. In percentage terms, the computer category was the fastest grower up more than 75% sequentially, driven primarily by inbox fast chargers for tablets, reflecting work from home and learned from home demand. Communication revenues were up about 25% sequentially, driven by continued strength in fast charging for smartphones. Consumer revenues were up mid-teens sequentially on improved demand for major appliances, offset by seasonal softness in air conditioning. Industrial revenues were down high single digits as growth in home automation and power tools were offset by weakness in broad-based industrial application as well as high-power, where a number of projects have been slowed down by the pandemic. Revenue mix for the quarter was 32% communication, 31% consumer, 28% industrial and 9% computer. As expected, revenue mix shifted towards low-margin categories resulting in an 80 basis point reduction in non-GAAP gross margin to 50.3%. Non-GAAP operating expenses were $35.9 million, in line with our guidance. Other income for the quarter was $0.9 million, down from the prior quarter as expected due to the lower interest rate environment. The non-GAAP effective tax rate for the quarter was just under 7%, resulting in non-GAAP earnings of $0.40 per diluted share. Cash flow from operations were $16.2 million, down from the prior quarter due to working capital fluctuations with receivables rising as a result of back-end loaded shipments during the quarter. Capital expenditures were $14.1 million driven mainly by building construction and capacity additions, including continued investment in GaN capacity. We paid out $6.6 million in dividends following the dividend increase we announced last quarter in conjunction with our stock split. Cash and investments on the balance sheet declined by $2 million from the prior quarter. Internal inventories were 155 days at quarter end, down 23 days from the prior quarter. We continue to maintain an above normal level of inventory, given the uncertainty of the supply and demand environment. But I do expect inventory days to continue to glide downward and ultimately return to our target range by the second half of 2021. Channel inventories fell sharply during the quarter, ending September at 4.3 weeks, down three weeks compared to the prior quarter driven by the strong sell-through that Balu mentioned, particularly in cellphone and appliance applications. Looking ahead to the fourth quarter, we expect revenues to be in the range of $130 million, plus or minus $5 million, with the sequential increase driven mainly by continued strength in fast charging as well as broader channel restocking. Mix should tilt a bit further in the direction of communication, resulting in a slightly lower gross margin compared to the third quarter. Specifically, I expect non-GAAP gross margin to be approximately 50%. Turning to expenses, while we did not reduce headcount in the response of the pandemic, we did defer a portion of our hiring to the later part of the year, and we will begin to see the impact of that in our OpEx in Q4. As a result, non-GAAP OpEx should increase modestly to around $37 million. That puts us on track for just a slight increase in full year expenses, despite increased headcount and despite giving normal salary increases in April. Naturally, the modest growth in expenses this year does reflect savings from the deferred hiring and from reduced travel and events. With a significant number of new hires coming on board in Q4 and a rebound in spending on travel and events, OpEx will likely grow at an above normal percentage rate in 2021. Other income, which is driven mainly by interest income, will continue to trend downward, reflecting the lower interest rate environment. Specifically, I expect other income to be around $800,000 in Q4 and to continue to taper down in subsequent quarters. Finally, the non-GAAP effective tax rate for Q4 should remain around 7%. And with that, I'll turn it back over to Joe.

Thanks Sandeep. We'll begin the Q&A session now. Operator, would you give the instructions for the Q&A.

[Operator Instructions] Our first question comes from the line of Ross Seymore of Deutsche Bank. Your line is open.

Hi guys, congratulations on the strong results and the strong guidance. Balu, I just wanted to talk a little bit about the filling of the channel that you're talking about in the fourth quarter. Obviously, it got much leaner in the third quarter. It's going to refill in the fourth quarter, but as you look into the first quarter? How are you expecting seasonality to behave? And is seasonality even a useful framework these days, given the change in your mix and then the COVID related impacts across the seasonality earlier this year?

Ross thanks for the compliment. Yes, it's a very good question. Normally, Q1 will be flat to slightly down. But based on all of the drivers we have, especially in communications, and also, we think the appliances it will come back. We see strong sell-through in appliances. We think the Q1 could be actually slightly higher because of that. So it is very different from normal seasonality, I would agree with you.

The other thing Ross, I think you have to - it's a little early. We'll have to see how Q4 sell-through happens and how much of the Q3 bookings ends up into the real demand that which you know was very strong and how much is the end customer rebuilding. Even though the channel inventories could go up, but I don't think they're going to go back to the levels they were in this coming quarter itself. It will take couple of quarters to get there.

Got it. And I guess as my follow-up, I just want to think a little longer term. Sandeep, you talked about OpEx growth being a little higher in 2021. I think that makes total sense, given the lower base for all the COVID related reasons in 2020. But if we thought about that relative to the revenue growth, and I know you're not guiding for next year on revenues as a whole? But if you did the puts and the takes relative revenue growth versus OpEx growth, any sort of framework that you can provide to us to give a little more like boundaries around that?

Absolutely as you know, our model is low double-digit. And as Balu indicated, we are kind of right there for this year. Added to that, I think the COVID-related spending as well as like travel, sales conference. I think this year's expenses are about abnormally low by about $3 million. So, if you take the guidance that we have given, that will give you somewhere around $142 million to $143 million for the year non-GAAP. If you add the $3 million on top of that to normalize this year and then give it a 7% to 8% growth, that will give you a number of somewhere around $155 million, $156 million. So really, I'm trying to talk about normalizing this year, and then you'll get a 7% to 8% growth. And that's slightly above our model. If you - without giving guidance for next year, if you believe our model of low double digit.

That's perfect, thank you.

Thanks Ross.

And our next question comes from the line of Tore Svanberg from Stifel, Nicolaus. Your line is open.

Yes thank you and congratulations on the results. Balu, you mentioned 80 customers and 12 applications now using GaN that's pretty remarkable. Are you really starting to see GaN penetrate non-smartphone applications faster than expected?

I won't say faster than expected because the cycle times are longer in other areas. But I would say that I am very, very enthused by the level of design activity that's going on. And we have won in applications that I would have never thought would be a good candidate like TVs and refrigeration - refrigerators, air conditioning for different reasons. In TVs because TVs have much higher efficiency requirements, plus the hotspot is a big issue on TV. And they really like the fact that we don't need a heatsink because we don't generate very much heat at all on the chip. So you don't get any hotspots on the back of the screen. And then if you look at refrigerators, it's not the size. Obviously, refrigerator doesn't care as much about the size of the power supply. But the lack of heatsink is a huge advantage for them. They say that heatsink is a major reliability problem, especially when it - these are shipped due to vibration, sometimes the heatsinks fall off or they break the printer circuit board. And they are very, very excited about not having a heatsink. And also, the efficiency also is becoming important with the tighter regulations in that area. In air conditioning, there are very tight regulations and efficiency, and that's really pulling our InnoSwitch GaN, InnoSwitch devices. The other one that is not so surprising is the USB wall outlet. The USB PD can go up to very high-power levels. They can go up to 100 watts. And trying to fit that in the wall outlet where there is no ventilation, means that you have a very high efficiency, very little heat, and there is no room to put a heatsink. And so that's a perfect application. That comes under - home and building automation, which is in the industrial part of our market. So we are seeing areas that we never expected. The other one is Compact Ballast. These ballasts, again, go into the - above the ceiling. And they are concerned about heat, they're concerned about weight, they're concerned about the compactness and most of all, efficiency. And so, we are seeing a nice application in lighting, for commercial lighting. So GaN seems to be an excellent fit for literally the entire market - the one I forgot to mention, the most important one is Notebooks. Notebooks are typically 65 watts, and you know how big those old chargers are. Now that they see really small chargers for cellphones and tablets, they don't want to have very large chargers anymore. So we are working with a number of Notebook customers and [won] designs, as we mentioned in Notebook, and that's another area we will do very well. And I saw that tablets, we did very well. As a result, we grew a computer market by 75% sequentially.

Yes, thank you for that detail. And I had a question on the fast charging business for communications or smartphones. Could you talk a little bit about the dynamics there, especially on pricing, right? Because if the adapters are in the box, I would assume there's heavy pricing pressure. But if this is - it looks like it's going to be more of an accessory market. Does that mean that your pricing is going to be quite a bit better?

Well, the volumes will be smaller, and therefore, the pricing will be better. And the other thing you have to remember is that the accessory market is growing very fast. They want to take advantage of the fact that at least one major OEM has gone out of the box. And if you noticed in their website, they also offer third-party chargers that use our product, especially GaN product to make it really small and attractive. The other trend we are seeing is that multiport adapters. Where you have more than one port, and which means we have more than one chip. Each port typically has one of our InnoSwitches. But as you go to higher power levels, because multiport, by definition is higher power level. You had to have at least 30 or 40 watts per port. The GaN becomes a very attractive option to keep the size down. And you'll see number of multiport adapters using our GaN technology. So we believe that the GaN. The out-of-the-box type of trend will encourage GaN and therefore, higher ASP and higher dollar content and higher gross margin content for us. But again, that strategy will be different for each OEM because some of the OEMs, especially the Chinese OEMs they really sell based on their ability to charge very fast. They are constantly increasing their power levels to charge their phones faster and faster. They're already at 65 watts, and they're talking about 100 watts to charge their phone even faster. And in their case, because it's a selling point, they're unlikely to switch to an out-of-the-box model, at least in the near term. Eventually, it might happen, but in the near term, it's unlikely to happen. And also in their case, they use proprietary protocols. So they're not using USB PD. They're using their own protocols. They believe that allows them to charge faster. So they have a different strategy in terms of charging. So I think, we are in a fantastic position in this market. We do anticipate in the long run, many of them will go out of the box, but that has both pluses and minuses, even though the volume is lower. It will allow us to sell the higher end of our products, which is GaN products with higher power level, higher ASP and more gross margin dollar content.

Very good. Just one last question, could you - as I know this is a longer-term opportunity for you. But can you just give an update on your progress on the automotive market?

Well, the automotive market, as we mentioned earlier, will take several years to develop, but it depends on where you go into the automotive. The drivetrain is the slowest because it has to go through a lot of safety approval and the field test and so on. But there are other applications where our current products like InnoSwitch, LinkSwitch, those can be used immediately. And we just recently introduced a number of products, including our Qspeed diodes for automotive. And we'll have a tiny little bit of revenue this year. And you'll see gradual increase in this revenue over the next several years, but the drivetrain itself will take a few years before we see significant revenue. But in the long run, it could be our largest addressable market. And the electric vehicle market is supposed to take-up in about three to four years. I think we are well positioned to get a good share of that, and we are working towards a number of products to go after that market.

Thank you.

Thanks Tore.

And your next question comes from the line of Karl Ackerman from Cowen. Your line is open.

Yes, hi thank you good afternoon. Two questions, if I may one of your larger peers for high-voltage power modules announced a partnership this week that seek to address high voltage, industrial and automotive applications. I'm curious does this change your view on the competitive landscape or share gain potential across or appliance in industrial business? And if not, of those 80 customers using GaN, you referenced. What mix of those are within industrial and appliance that appear to augur well for your opportunity in 2021 and 2022?

Hi Karl, nice to meet you for the first time, I don't think we have talked before and thank you for the question. I assume you're referring to the ST press release where they are cooperating with the Sanken on the IPM for 3 kilowatts and above.

That's right.

First of all, we don't - yes, we don't play in that market. Did you confirm that, Karl?

Correct.

Okay. So first of all, we don't play at that power level. We do have - we do address motor control, but at much lower power levels, below 400 watts with our BridgeSwitch. So it really doesn't directly impact us. And having said that, we compete regularly with Sanken and ST very well I mean, not at that power level, but at lower power levels. And we have been gaining share against Sanken in appliances and industrial applications consistently. We have a very high share, I mean in the 40% to 50% share of the appliance market. And a lot of that has come from Sanken. And again in the industrial market, we have gained share against ST over many, many years. So, we think we have very compelling products. And so it really doesn't affect our outlook going forward as far as this announcement is concerned. You had another question that is where do we fit in the industrial market with InnoSwitch GaN. Our biggest opportunity, near-term opportunity is the USB wall outlet. These are USB outlets right next to your power outlet on the wall. And there we bring significant advantages. I actually forgot to mention one other important advantage in that application that is standby consumption. When you put a power supply in the outlet, you have no way of disconnecting it from the grid, which means that if you have this 20 or 30 of them installed in your house. They are constantly drawing some level of power even when you are not using them. So it's really important you have the lowest power level possible. And we have a very low standby consumption with InnoSwitch and InnoSwitch GaN, and that's a huge attraction in that market. Having said that, I believe there are other applications that we will get into overtime. We already see possibilities in other areas in industrial. But our near-term exposure with GaN is the USB PD wall outlet. In terms of the percentage, I would say predominantly, we are getting GaN revenue this year from mobile applications including Notebook adapters and the tablets and so on and multiport chargers for all three of them. That's the cellphones, tablets and Notebooks. And so the appliance and - basically the consumer, which is appliances and industrial market will gradually come up because they are much longer design cycles. But we are really surprised at the level of design activity we are seeing and the acceptance of GaN in those markets.

Understood, thank you for that color. If I can go back to mobile, I think one of the investor concerns is that there will be perhaps this demand air pocket is clearly the largest handset OEM and the U.S. is no longer selling the charger with the initial sale of the phone. But what indications are you hearing from your channel partners about the potential for an upgrade cycle of USB-C chargers? Second, perhaps maybe a 2021 story, you spoke in the past about design wins at China handset OEMs, but could you talk about whether the fast charging growth opportunity for you will percolate into the low and mid-range 5G handsets next year or will the opportunity be tied predominantly to flagship models? Thank you.

Good question, Karl. So the fast charging, fast chargers are definitely penetrating to the lower end phones. The China customers are aggressively pushing fast charging all the way down to their low-end phones. And that's one of the reasons our revenue is growing. It will continue to grow in 2021. The other reason is our Chinese customers, especially - and in fact, all our companies even non-Chinese OEMs in cellphones are expecting to gain share from the challenges that Huawei has in terms of being able to ship phones. And so they're all preparing to gain share from that, and that's adding to the demand, that we already have. So we are not only penetrating more, but we're also - our customers are also gaining share from Huawei. The third aspect is we have gained significant share over the last year in terms of new design wins at all of our OEMs. So that is compounding the growth rate. Now having said all of that, I would say that the recent - the dramatic increase in bookings is a result of multiple things. One is - the three items I mentioned, but also, I think the distributors have a needed reaction when they see a huge sell-through, they place a lot of orders. So we have to - make some judgments on that. And we also don't know how exactly the Huawei opportunity will benefit each and every one of our OEMs, but we have a pretty good understanding of what the total volume that's at stake for them. So we are able to gauge at least overall, what that opportunity is for all of us. So there are many, many different things that are helping us simply because we are in a perfect position with our products and technology that is InnoSwitch and GaN in all of these markets. And actually, I'll add one more. Even the one that announced that they're going to go out of the box, the advantage is that when it goes out of the box and a consumer buys a phone, even if they buy a low-end phone, the chances are they're going to buy - the fast charger. Because when you're paying $400, $500 for a phone, nobody minds paying $19.99 for a really high-end charger rather than paying the same amount for a 5-watt charger because you get 4 times the charge rate. And also the wireless charging that has been announced will also require the same fast charger, the 20-watt charger. So in some sense, the going out of the box gives us the exposure to the entire product line, not just the high end, like we have seen in the past. Now it is up to the consumer to choose what type of charges they want. And so, we'll have to see what attach rate would be. But we are optimistic the attach rate will be reasonable because most consumers would prefer to have fast chargers. And we talked to a number of us - our phones have talked to friends, they would very much like to have a fast charger. And the interesting thing is most of the phones, even going back three or four years, can handle fast charging. It's just that for cost reasons, they are shipping with the lowest cost charger, with a 5-watt charger. Now it's up to the consumer to make the decision. So - and from an OEM perspective, it also makes sense because now they can make - by going to an accessory model they can make the charger much more attractive in terms of size and multiport features. And also, they can make more margins. It's no longer adding to the cost of the phone.

Thank you.

You're welcome.

And our next question comes from the line of Christopher Rolland of SIG. Your line is open.

David Haberle on behalf of Chris Rolland thanks for taking our questions and congratulations on the solid results and guidance that you produced here. My first question, I wanted to ask to clarify on the computing strength here. Is this more a product of work-from-home driving tablet strength this quarter and next quarter or is this the GaN revenue really starting to trickle in here? I guess the question is, you guys have been running at mid single-digits per quarter in that computer market for a while now, is low double-digits, a new possible run rate going forward?

Well, it all depends upon how the rest of the other markets grow. But I would say that we expect the computer market to continue to grow next year. And it really comes from our product getting in the box on tablets. And so, it's not - it's a combination of two things. One is we are in the box. And the second one is the tablets are doing extremely well, thanks to work-from-home and learn-from-home situations right now. So the combination of those two is giving a huge boost. But there are also other areas we expect to grow, one is Notebooks. I just mentioned that we have won some designs in Notebooks. We haven't seen the revenues yet, and that would help us next year. The other area is we introduced a product called InnoMux that goes into monitors. And we have won some designs that will add to revenue. We already started adding to revenue, but next year, it will be more. You'll see the full year impact of those revenues and that will increase over time. The other area, we have seen a small increase is the standby power supplies in desktops, which we - which has been declining over time because people don't buy desktops as much anymore. But we are seeing an increase in that volume because they're selling more desktops in other countries, third world countries.

Got it, thank you for the color there, a lot of good things going on in that market. If I could just ask the second one on industrial here, I think the high-power market kind of been weighing for a few quarters in a row now? I think infrastructure projects is understandable with the pandemic going on at these have kind of hit a standstill, but do you have any visibility on when these projects start to reaccelerate and kind of get this industrial business going again for next year?

That's a very good question. The good news is we still won the designs. The designs are still there. It's just a question of when the governments decide to trigger these projects. I don't know an exact date, but I have to believe they have to add more energy capacity, especially in developing countries like India, China, where they have continuous increase in [electricity] demand and work-from-home probably makes it worse in some ways. So it's just a question of time. I mean there is a lot of pent-up requirement to add wind capacity, it's all renewable energy, including solar panels, but also grid level installations, high-voltage DC grid and the last one is traction like locomotives. China has done very well in locomotives. India has a very aggressive plan to add electric locomotives. They are trying to convert all diesel engines into electric. And so, there is a long-term plan. And that was supposed to kick off this year, but again, that's been on hold. It's really hard to say. Part of me says maybe the COVID situation gets better, those will get triggered. But I have no way of knowing. It's just up to the government. But the good news is it doesn't change our position. We have the share, we have won the designs. It's just a question of when they start implementing them.

Thank you.

[Operator Instructions] Your next question comes from the line of Gus from Northland Capital. Your line is open

Just real quick, what was CapEx in the quarter?

For the quarter, we spent about $14 million.

Okay. And in the past, you had - you bought equipment, I think, for your fabs and you also buy your own tester equipment for your OSATs. Can you give me a split of what they were?

So, I think the way to look at it, including this year - we're going to probably spend as I had indicated earlier in the year, that this will be a year of a little abnormal spend. Where we're going to spend about $55 million in CapEx, $15 million is for buildings and about $40 million is on equipment, including GaN, and quite a bit is on the back end, too. However, looking ahead, this will get normalized back. Where we will normally go back into the equipment into the $20 million to $25 million for next year in 2021 but we have another $10 million to $15 million, more to spend on our building and solar that we have planned for next year. But if you want to look at it ongoing in a long-term, 5% to 6% of revenue is a regular model.

Got it. And then given your spending on [silicon] and GaN, I would - is that you're comfortable with your book to ramp capacity for these products?

Yes, we have got adequate capacity related to that. And that's something that we have been investing and will continue to invest. Because this is a great opportunity for us - as Balu had indicated, with all the design wins we are seeing in the future years, GaN is a great opportunity. So we are strategically investing here.

Yes, I'd like to add that I truly believe GaN is going to replace silicon about certain power level, like roughly about 30 watts and above. GaN makes a lot of sense. And so, we are really investing in GaN. We have already built enough capacity in the near term. But as time goes on, we can add more capacity. And we think we are in a very strong position here, way ahead up for all of our competitors in GaN. As far as we know, we are the only one shipping GaN in volume. And we really want to take advantage of that. And all - almost all of our new products are based on GaN about a certain power level.

Got it, got it. And you answered my next question, but you're not seeing anybody else in the marketplace with GaN products?

Well, there are a lot of people talking about it, sampling but we don't see anybody in high-volume production at all.

Got it, got it. And I'm sorry, this is sort of a net question. Are you guys building on 8-inch wafers at this point or is it still 6?

Are you talking about silicon or GaN?

GaN.

GaN, we can't disclose that. We don't disclose our technology. It's to too sensitive and it's a competitive situation.

Okay fair enough. And then just in terms of growth in compute stocking, compute next year, is that growth going to be split between tablets and notebooks or do, you have any color there?

Well, next year the communications, which is primarily cellphones, will continue to drive growth in revenue. But we also expect the consumer market, which is appliances, including air conditioning, to come back next year. So that will be another area of growth for us. Industrial, again it's mixed. Some areas will grow very nicely like HBA, which is home and business automation and tools and battery-operated tools. Those we expect to continue to grow very well next year. What we don't know is how the high-power is going to do. We talked about it earlier, whether it will come back next year or will the infrastructure programs will be delayed further. So, depending on what that happens, the industrial may or may not grow because it's just a question of whether the HBA and the tools will be able to overcome any shortcomings of high power that is still a question mark. And then when it comes to computer, there are many puts and takes. We obviously had a very strong quarter in tablets, and we'll continue to have in Q4. This - is definitely driven by work-at-home and learn-at-home type of situation. And that may slow down a little bit although you could say there may be holiday sales. So it's a little bit hard to predict. If I were to take a guess, I would think computer will also grow slightly next year. The big question mark is industrial.

I think we are really positioned very well. As we have said that we have drivers in each market. And I really believe next year would be a good year for all end applications directionally for us.

Got it all right thanks so much.

You're welcome.

And thank you everyone, for your participation in today's Q&A session. I'll now turn the call back to Mr. Joe Shiffler for his closing remarks.

All right, thanks, everyone. We'll leave it there. There will be a replay of this call available on our website, our investor website, investors.power.com. Thanks again for listening, and good afternoon.

Thank you.

And this concludes today's conference call. You may now disconnect.